Hedgehog proteins act as morphogens in a wide variety of tissues during embryonic development (Ingham, 1995; Perrimon, 1995; Johnson and Tabin, 1997; Hammerschmidt et al., 1997). Vertebrate hedgehogs are crucial to a number of epithelial-mesenchymal inductive interactions during neuronal development, limb development, lung, bone, hair follicle and gut formation (Ericson et al., 1995; Roberts et al., 1995; Apelqvist et al., 1997; Ericson et al., 1997; Hammerschmidt et al., 1997; Johnson and Tabin, 1995; Pepicelli et al., 1998; Litingtung et al., 1998; Roberts et al., 1998; Dodd et al., 1998; Dockter, 2000). Mammalian hedgehog genes consist of sonic, indian and desert which are highly conserved between species (Zardoya, 1996). Sonic hedgehog (shh) is expressed widely during development and sonic null mice are embryonic lethal with multiple defects beginning early to midgestation (Bitgood and McMahon, 1995; Chiang et al., 1996; Litingtung et al., 1998; St-Jacques et al., 1998). Indian hedgehog (ihh) is expressed less widely and indian null mice survive till late gestation. However, Ihh null mice exhibit severe stunting of skeletal growth which correlates to the role of Ihh in regulating bone growth plate (St-Jacques et al., 1999; Karp et al., 2000). Desert hedgehog (dhh) is the most restricted in expression and Dhh null mice are viable, but as expected from the expression pattern, male gonads do not develop completely and the peripheral nerves develop in a disorganized fashion (Bitgood et al., 1996; Parmantier et al., 1999).
Hedgehog signalling occurs through the interaction of hedgehog protein with the hedgehog receptor, patched (Ptc) and this interaction's modulation of the co-receptor smoothened (Smo). The mammalian genome contains 2 patched genes, ptc1 and ptc2, both of which encode 12 transmembrane proteins containing a sterol sensing domain (Motoyama et al, 1998; Carpenter et al, 1998). The interaction of Hh and Ptc inactivates the repression of smoothened (Smo), a 7 transmembrane protein which then leads to activation of fused (Fu), a serine-threonine kinase, and the disassociation of a transcription factor, Gli, from the microtuble-associated Fu-Gli-Su(fu) complex. The uncomplexed Gli protein is transported to the nucleus where it activates downstream target genes of the hedgehog pathway including the ptc1 and gli1 genes (Ding et al., 1999; Murone et al, 1999a; Murone et al, 1999b; Pearse et al., 1999; Stone et al., 1999; Hynes et al, 2000).
Hedgehog genes have so far not been implicated directly in embryonic or adult angiogenesis. No vascular defects have been reported in shh, ihh or dhh knockout mice. However, we show here that cells in the adult vasculature both express ptc1 and can respond to exogenous hedgehog and, more importantly, hedgehog is able to induce robust neovascularization in the corneal pocket model of angiogenesis. The angiogenic response to hedgehog appears to occur through the activation of mesenchymal cells to produce VEGFs and Angiopoietins.
Angiogenesis, the process of sprouting new blood vessels from existing vasculature and arteriogenesis, the remodeling of small vessels into larger conduit vessels are both physiologically important aspects of vascular growth in adult tissues (Klagsbrun and D'Amore, 1991; Folkman and Shing, 1992; Beck and D'Amore, 1997; Yancopoulos et al., 1998; Buschman and Schaper, 2000). These processes of vascular growth are required for beneficial processes such as tissue repair, wound healing, recovery from tissue ischemia and menstrual cycling. They are also required for the development of pathological conditions such as the growth of neoplasias, diabetic retinopathy, rheumatoid arthritis, psoriasis, certain forms of macular degeneration, and certain inflammatory pathologies (Cherrington et al., 2000).
The ability to stimulate vascular growth has potential utility for treatment of ischemia-induced pathologies such as myocardial infarction, coronary artery disease, peripheral vascular disease, and stroke. The sprouting of new vessels and/or the expansion of small vessels in ischemic tissues prevents ischemic tissue death and induces tissue repair. Certain growth factors such as those in the vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) families are able to stimulate vascular growth by acting on endothelial cells to induce angiogenesis. Other factors have also been shown to have angiogenic and arteriogenic activities such as MCPI (Buschman and Schaper, 2000) and angiopoietins. In preclinical models of myocardial infarction, both FGFs and VEGFs have been able to improve myocardial revascularization and function (Yanagisawa-Miwa et al, 1992; Battler et al., 1993; Harada et al., 1994; Banai et al., 1994; Unger et al., 1994; Mesri et al., 1995; Pearlman et al., 1995; Landau et al, 1995; Lazarous et al., 1996; Engler, 1996; Magovern et al., 1997; Shou et al., 1997). Also in models of peripheral vascular disease, VEGF and other angiogenic factors are able to induce angiogenesis and improve vascular perfusion of the ischemic limb (Majesky, 2000; Takeshita et al, 1996 and 1994; Rivard et al., 1998 and 1999, Isner et al, 1996).
A number of these factors are also implicated in vascular growth in pathological conditions such as tumor expansion, diabetic retinopathy and rheumatoid arthritis. The inhibition of vascular growth in these contexts has also shown beneficial effects in preclinical animal models (Klohs and Hamby, 1999; Zhu and Witte, 1999; Cherrington et al., 2000). For example, inhibition of angiogenesis by blocking vascular endothelial growth factor or its receptor has resulted in inhibition of tumor growth and in retinopathy (Fong et al., 1999; Wood et al., 2000; Ozaki et al., 2000). Also, the development of pathological pannus tissue in rheumatoid arthritis involves angiogenesis and can be blocked by inhibitors of angiogenesis (Peacock et al., 1995; Storgard et al., 1999).
Thus, the induction of angiogenesis and vascular growth is beneficial for tissue repair and would healing whereas inhibition of angiogenic growth factors can prevent angiogenesis driven pathologies. It would be useful to develop novel therapeutics that modulate angiogenesis.
Furthermore, the present invention demonstrates that the inhibition of hedgehog signaling decreases unwanted cellular proliferation, growth, differentiation and/or survival. Such unwanted cellular behavior is often observed in tumor cells (e.g., in cancer). Accordingly, the present invention provides methods and compositions of decreasing unwanted cellular behavior, such as occurs in cancer, by administering hedgehog antagonists.